Updated October 2026 · By Farrukh Moin Ahmed, digital marketing expert
Short answer: On prospecting vs retargeting, the answer depends on budget. On a small budget, don’t split them: run one all-funnel purchase campaign so Meta learns from all your purchases. Once each stage has enough purchases to learn on its own, separate them, size retargeting to your warm audience and frequency, give prospecting the rest, and judge each stage on its own job.
Key takeaways
- Retargeting usually shows a higher ROAS because its audience is already warm, not because it deserves more budget.
- Under about 50 purchases a week, combine both stages in one all-funnel campaign.
- When separated, size retargeting to its audience and frequency, not a fixed ratio.
- Judge prospecting on new customers and retargeting on cost per purchase.
- Use total store revenue against total ad spend as the final check.
What are prospecting and retargeting in Meta Ads?
Prospecting means showing ads to people who haven’t interacted with your brand yet. They haven’t visited your site, engaged with your Instagram or bought before. The job is to introduce the brand and find new customers.
Retargeting (also called remarketing) means showing ads to people who already have: site visitors, people who added to cart, video viewers, Instagram engagers, past customers. The job is to bring warm people back and turn interest into a purchase.
For years, the standard Meta setup for an online store was one campaign for each, with a budget split between them. That’s still a valid setup for some accounts. But Meta’s delivery system has changed a lot, and for many stores today the right split is no split at all. This guide covers how to decide, and how to judge each stage if you do separate them.
Why does retargeting always look better than prospecting?
Open almost any ecommerce Meta account and retargeting will show a higher ROAS than prospecting. It’s tempting to move budget toward the better number. That’s usually a mistake, for three reasons:
- Many retargeted buyers would have bought anyway. Someone who added to cart yesterday is already close to buying. If they see a retargeting ad and then buy, Meta credits the ad, even if they’d have returned on their own.
- Retargeting harvests what prospecting planted. Every person in a retargeting audience got there somehow, often from a prospecting ad. Cut prospecting and the retargeting pool shrinks, and the impressive ROAS goes with it, a few weeks later.
- View-through credit flatters retargeting. Warm audiences see your ads often. With a 1-day view attribution window, an ad that was merely scrolled past can still claim a sale.
So a higher retargeting ROAS doesn’t prove retargeting deserves more money. It mostly proves retargeting audiences are warmer.
Should prospecting and retargeting be separate campaigns?
This is the decision that matters most, and it comes down to purchase volume. Meta’s delivery system learns from optimization events. Its learning phase is generally described as needing around 50 optimization events within seven days per ad set. Split a budget into separate campaigns and each one gets only a slice of those events.
| Weekly purchases from Meta | Recommended structure |
|---|---|
| Under about 50 | One all-funnel purchase campaign. Meta decides who sees what. |
| About 50 to 100 | One main campaign; consider a small separate retargeting campaign only if warm audiences are large. |
| Comfortably over 50 per stage | Separate prospecting and retargeting, each judged on its own job. |
Meta’s own tools have moved in this direction. Advantage+ sales campaigns, for example, deliver to new and existing customers in one campaign, with an option to cap how much goes to existing customers. That reflects how delivery works now: one system with plenty of data usually allocates better than several systems with a little each.
What happened when Woda Bags combined them?
Woda Bags, a US bag brand, is a clear example. Before I took over the account in late April 2026, it ran three campaigns: a prospecting traffic campaign, a retargeting-only purchase campaign and a combined prospecting-plus-retargeting purchase campaign.
| Campaign | Spend | Purchases | Cost per purchase | ROAS |
|---|---|---|---|---|
| Retargeting only | $791 | 9 | $87.87 | 2.18 |
| Prospecting + retargeting | $6,569 | 141 | $46.59 | 4.59 |
| All-funnel (since late April 2026) | $1,053 | 41 | $25.69 | 10.01 |
Meta-reported, 7-day click and 1-day view attribution. The campaigns ran in different periods and at different budgets, so this isn’t a controlled test.
The interesting line is the retargeting-only campaign. Conventional wisdom says it should be the most efficient campaign in the account. It was the least efficient purchase campaign, because a small brand’s retargeting audience is small, so the campaign saw very few purchases each week and couldn’t learn. The full breakdown, including the traffic campaign, is in Meta Ads campaign structure: one campaign or three? and the Woda Bags case study.
If you do separate them, how should you split the budget?
Once each stage can stand on its own data, the split should follow audience size rather than a fixed percentage.
- Size retargeting to its audience. Retargeting can only spend usefully on as many people as you bring to the site. If your warm audience is 20,000 people, a large retargeting budget mostly raises frequency, showing the same people the same ads again and again.
- Watch frequency as the limit. When retargeting frequency climbs and cost per purchase rises with it, the audience is saturated. Extra budget belongs in prospecting.
- Give the rest to prospecting. Prospecting is where growth comes from. In most growing accounts it takes the large majority of spend, because it’s the only stage that can expand the audience.
- Exclude recent buyers from prospecting. Otherwise prospecting takes credit for existing customers and inflates its numbers.
Some practitioners quote rules like 80/20 or 70/30 in favour of prospecting. They’re reasonable starting points, but audience size and frequency should overrule any ratio.
How should you judge each stage?
Judging both stages on ROAS alone will always push budget toward retargeting. Give each stage its own scorecard instead:
| Prospecting | Retargeting | |
|---|---|---|
| Main job | Find new customers | Convert warm interest |
| Primary metric | Cost per new customer | Cost per purchase |
| Supporting metrics | Reach, new site visitors, growth of warm audiences | Frequency, conversion rate |
| Warning sign | Cheap clicks that never add to cart | Rising frequency with rising cost per purchase |
| Store-level check | Are new-customer orders growing? | Is total revenue growing, or just being re-credited? |
The store-level check matters most. Compare total store revenue to total ad spend across every channel. If retargeting spend goes up but total revenue doesn’t, retargeting was taking credit for sales you’d have had anyway. Reliable purchase data underpins all of this, which is why I set up the Pixel and the Conversions API before changing budgets. More on that in analytics and tracking.
What creative belongs in each stage?
Prospecting and retargeting audiences are asking different questions, so they need different ads.
- Prospecting creative answers “what is this and why should I care?” Product in use, a clear benefit, a reason to stop scrolling. Most creative testing belongs here, because this is the audience that doesn’t know you.
- Retargeting creative answers “why should I buy now?” Reviews, shipping and returns, sizing, a new colour, a time-limited offer if you run them. Catalog ads, which show the exact products someone viewed, cover much of this automatically.
In a combined all-funnel campaign, include both types in the same campaign. Meta will show the reassurance ads more to warm people and the introduction ads more to cold people, without you managing separate budgets.
Which audiences belong in each stage?
If your volume justifies separate campaigns, define the stages by audience so they don’t overlap:
| Stage | Include | Exclude |
|---|---|---|
| Prospecting | Broad targeting, or broad with light suggestions; lookalikes of purchasers if you use them | Site visitors (last 30 to 180 days), engagers, purchasers |
| Retargeting | Site visitors, add-to-carts, checkout starters, video viewers, Instagram and Facebook engagers | Recent purchasers (match your typical repurchase cycle) |
| Existing customers (optional) | Past purchasers, for new products or replenishment | Very recent purchasers |
Keep retargeting windows long enough to hold a useful audience. A 7-day window on a small store may contain a few hundred people, which is too few for Meta to deliver efficiently. A 30 to 90 day window is often more workable, with the most recent visitors weighted through creative rather than separate ad sets.
Whatever the setup, the exclusions only work if the Pixel and Conversions API are recording visits and purchases accurately. Audiences built on patchy data leak between stages, and the numbers for each stage stop meaning anything.
What common budget mistakes should you avoid?
- Moving budget to retargeting because its ROAS is higher. This works for a few weeks, then the warm pool shrinks and both stages decline.
- Running prospecting on a traffic goal. Traffic campaigns find people who click, not people who buy. In Woda’s earlier account, a traffic campaign produced over 16,000 link clicks and no attributed purchases.
- Splitting a small budget three or four ways. Each campaign starves of data. Consolidate until volume allows separation.
- Not excluding purchasers. Retargeting recent buyers with the product they just bought wastes money and annoys customers.
- Ignoring the site. If product pages convert poorly, both stages look expensive. Fixing the page with conversion rate optimization improves every campaign at once.
How does Google Ads fit into this?
Meta isn’t the only retargeting channel. Google’s Performance Max and remarketing lists also reach past visitors, and branded search catches people who come back to look for you. That means some of the warm demand your Meta prospecting creates will be captured on Google instead. When you evaluate Meta prospecting, watch branded search volume and Google Shopping results too. How the two platforms divide the work is covered in Google Ads vs Meta Ads for ecommerce.
The short version
On a small budget, don’t split prospecting and retargeting at all. Run one all-funnel purchase campaign and let Meta allocate. Once each stage has enough purchases to learn on its own, separate them, size retargeting to its audience, give prospecting the rest, and judge each on its own job with store-level revenue as the final check.
For the wider approach, see Meta Ads for ecommerce, fashion and accessories and the ecommerce marketing hub. Want a second opinion on your Meta structure? Get in touch.
If you want a Facebook Ads specialist to look at how your budget is split, that is part of how I run Meta Ads for online stores.
Frequently asked questions
What percentage of Meta budget should go to retargeting?
There is no fixed percentage. Size retargeting to your warm audience and watch frequency. Many growing accounts end up with a minority of spend on retargeting and the rest on prospecting, but small accounts often do best with no separate retargeting at all.
Is retargeting still worth it on Meta?
Yes, but often not as a separate campaign. On small and mid-sized budgets, an all-funnel campaign reaches warm audiences anyway. Separate retargeting makes sense once the warm audience and purchase volume are large enough for it to learn on its own.
Why is my retargeting campaign expensive?
Usually because the audience is small, so the campaign gets few purchases and learns slowly, or because frequency is too high. Woda Bags’ retargeting-only campaign cost $87.87 per purchase before it was consolidated.
Do Advantage+ sales campaigns replace retargeting?
They can cover it. Advantage+ sales campaigns deliver to new and existing customers in one campaign and let you cap spend on existing customers, so many stores no longer need a separate retargeting campaign.
How do I know if prospecting is working?
Look beyond ROAS. Check cost per new customer, growth of your warm audiences, branded search volume and whether total new-customer orders in your store are rising.
Should I exclude past purchasers from prospecting?
Generally yes, so prospecting is judged on new customers. Some stores retarget past purchasers separately with new products or replenishment offers.
